1. Field of the Invention
The present invention relates to a video decoding method and a video decoder which is to decode a compressed video bit stream based on a single synchronization signal, and a digital TV system using the video decoding method and the video decoder.
2. Discussion of Related Art
Image compression technology along with digitalization are essential factors in assisting multimedia. Recently data digitalization is making rapid progress, and image compression technology is highlighted.
Moving pictures expert group (MPEG) is the standard of compression and encoding method of digital moving picture which is one of the most important factor in multimedia.
Various technological fields use MPEG technology as compressing algorithm incluiding communication, broadcasting, game, etc. Digital TV is one of them.
MPEG video compression algorithm is based on two basic technologies. It uses motion estimation and compensation in block units to reduce temporal redundancy, and uses discrete cosine transform (DCT) to reduce spatial redundancy.
DCT signals are transmitted through quantization process that reduces data by displaying data in limited bits, variable length coding (VLC) process that reduces total bits by displaying frequently-appearing value in bits of small number, and rarely-appearing value in bits of large number.
And motion vector is acquired in 16.times.16 block unit, and transmitted with the above VLCed coefficient information. The above motion vector is sometimes VLCed for the maximum efficiency of encoding.
Digital TV that uses this MPEG technology include tuner, demultiplexer, video decoder, and monitor. The tuner selects one-channel among multiple channels received from antenna and decode it, and demultiplexer selects one program which is included in one channel and separate it into packed audio and video bit stream. Because separated video bit stream is coded in variable rate, it is stored temporarily in video buffer for decoding, and output to video decoder in fixed rate. And video decoder removes overheads (all sorts of header information, start codes, and etc.) from input video bit stream, decodes into original value of pixels through VLD of pure data information, inverse-quantization process, inverse discrete cosine transform (IDCT), motion compensation process using motion vector, and sends the result to monitor as output.
The above motion compensation process uses forward and backward prediction blocks, and there are two kinds of motion compensation frame. P-frame is motion-compensated only by forward prediction, and is used to predict the next P-frame in itself. And it is also used in forward and backward prediction of B-frame (predicted frame in both directions). But B-frame is not used for prediction in itself.
That is to say, in MPEG algorithm, I (intra) picture is the base screen when any screen is compressed and encoded, and redundancy of spatial direction is removed through DCT and quantization of original signal per block. P (predicted) picture is the most predicted picture from I,P picture, and prediction error signal and motion vector between motion compensation and already encoded I-picture or P-picture are encoded per macro block. Concerning prediction error signal among these, quantization and VLC are sequentially executed after DCT operation per block. A B (bi-directional) picture is somewhat complex screen selected by method of selecting the most nearly predicted screen among the one consisted through prediction at previous I-, P-picture, backward-predicted one at I-, P-picture and average one of these two methods.
At this time, system time clock (STC), decoding time stamp (DTS) that presents the time when to decode each picture on the basis of STC, presentation time stamp (PTS) that presents the time when to display the decoded data on the basis of STC are multiplexed in A/V multiplex bit stream that enter into above demultiplexer. STC is the entire clock locked with the clock of encoder, and encoder and decoder have the same STC. Because video signal has delay internally, above encoder generates DTS and PTS on the basis of STC and transfer them together for A/V lip-synchronization and normal video decoding.
The above demultiplexer restores STC that make the basis of DTS,PTS from A/V bit stream and make it out to video decoder.
But the above video decoder has various problems.
First, it decodes the compressed video bit stream using DTS,PTS generated on the basis of STC, and displays it. And DTS,PTS signal intervals are not uniform because STC is 27 MHz and unstable clock that varies by surrounding conditions such as noises at the time of transmission. Accordingly, decoding and display become unstable, and especially, display in real system becomes unstable.
Second, circuit becomes complex because digital TV that uses the above video decoder is variously formatted, and synchronization are also input variously according to their format.
Third, the above video decoder uses 3 frame memory to display sequentially when frame is missed or repeated by various reasons. But this brings about waste of memory when display part that have various capability of displaying such as display format transformation is connected to rear side in the case of multi-format video, due to redundancy with a prime memory of display part.